home edit page issue tracker

This page pertains to UD version 2.

Treebank Statistics: UD_Italian-TWITTIRO: POS Tags: ADJ

There are 528 ADJ lemmas (10%), 669 ADJ types (10%) and 1237 ADJ tokens (4%). Out of 16 observed tags, the rank of ADJ is: 5 in number of lemmas, 4 in number of types and 10 in number of tokens.

The 10 most frequent ADJ lemmas: buono, nuovo, primo, bello, italiano, unico, pubblico, altro, grande, prossimo

The 10 most frequent ADJ types: buona, nuovo, primo, buon, bella, grande, italiana, pubblica, italiano, bel

The 10 most frequent ambiguous lemmas: buono (ADJ 186, NOUN 3), italiano (ADJ 30, NOUN 26), pubblico (ADJ 16, NOUN 1), altro (PRON 22, ADJ 15, DET 6), giusto (ADJ 11, ADV 3), solo (ADV 41, ADJ 10), vero (ADJ 10, ADV 1, INTJ 1, NOUN 1), caro (ADJ 7, NOUN 1), vecchio (ADJ 8, NOUN 2), solito (ADJ 7, NOUN 4)

The 10 most frequent ambiguous types: pubblica (ADJ 11, VERB 2), italiano (ADJ 9, NOUN 7), prima (ADV 13, ADJ 9, ADP 1), caro (ADJ 3, NOUN 1), giusto (ADJ 5, ADV 1), italiani (NOUN 18, ADJ 7), vero (ADJ 3, INTJ 1, NOUN 1), buono (ADJ 4, NOUN 3), altri (PRON 6, ADJ 4, DET 1), altro (PRON 10, ADJ 4, DET 3)

Morphology

The form / lemma ratio of ADJ is 1.267045 (the average of all parts of speech is 1.274961).

The 1st highest number of forms (7) was observed with the lemma “bello”: bei, bel, bell’, bella, belle, belli, bello.

The 2nd highest number of forms (5) was observed with the lemma “altro”: altra, altre, altri, altro, atri.

The 3rd highest number of forms (5) was observed with the lemma “buono”: buon, buona, buone, buoni, buono.

ADJ occurs with 4 features: Number (1203; 97% instances), Gender (981; 79% instances), NumType (53; 4% instances), Degree (10; 1% instances)

ADJ occurs with 8 feature-value pairs: Degree=Abs, Degree=Cmp, Gender=Fem, Gender=Masc, NumType=Card, NumType=Ord, Number=Plur, Number=Sing

ADJ occurs with 18 feature combinations. The most frequent feature combination is Gender=Fem|Number=Sing (374 tokens). Examples: buona, bella, italiana, pubblica, unica, igienica, nuova, piena, prossima, altra

Relations

ADJ nodes are attached to their parents using 21 different relations: amod (906; 73% instances), parataxis (82; 7% instances), root (74; 6% instances), xcomp (37; 3% instances), conj (35; 3% instances), nmod (19; 2% instances), obl (19; 2% instances), discourse (15; 1% instances), ccomp (13; 1% instances), advcl (12; 1% instances), flat:name (4; 0% instances), nsubj (4; 0% instances), obj (4; 0% instances), acl (3; 0% instances), advmod (3; 0% instances), parataxis:hashtag (2; 0% instances), dislocated (1; 0% instances), fixed (1; 0% instances), nsubj:pass (1; 0% instances), obl:agent (1; 0% instances), parataxis:appos (1; 0% instances)

Parents of ADJ nodes belong to 11 different parts of speech: NOUN (879; 71% instances), VERB (149; 12% instances), (74; 6% instances), PROPN (46; 4% instances), ADJ (41; 3% instances), SYM (21; 2% instances), PRON (17; 1% instances), ADV (4; 0% instances), X (3; 0% instances), NUM (2; 0% instances), AUX (1; 0% instances)

883 (71%) ADJ nodes are leaves.

123 (10%) ADJ nodes have one child.

60 (5%) ADJ nodes have two children.

171 (14%) ADJ nodes have three or more children.

The highest child degree of a ADJ node is 11.

Children of ADJ nodes are attached using 36 different relations: punct (195; 19% instances), cop (144; 14% instances), advmod (106; 10% instances), nsubj (86; 8% instances), obl (70; 7% instances), conj (57; 5% instances), parataxis (52; 5% instances), parataxis:hashtag (44; 4% instances), case (35; 3% instances), vocative:mention (34; 3% instances), cc (30; 3% instances), det (29; 3% instances), mark (29; 3% instances), advcl (21; 2% instances), nmod (20; 2% instances), discourse (19; 2% instances), dep (16; 2% instances), csubj (15; 1% instances), aux (7; 1% instances), ccomp (7; 1% instances), amod (6; 1% instances), discourse:emo (6; 1% instances), vocative (6; 1% instances), appos (3; 0% instances), nummod (3; 0% instances), list (2; 0% instances), parataxis:nsubj (2; 0% instances), acl (1; 0% instances), flat:name (1; 0% instances), nsubj:pass (1; 0% instances), obj (1; 0% instances), orphan (1; 0% instances), parataxis:appos (1; 0% instances), parataxis:discourse (1; 0% instances), parataxis:obj (1; 0% instances), xcomp (1; 0% instances)

Children of ADJ nodes belong to 16 different parts of speech: PUNCT (195; 19% instances), AUX (151; 14% instances), NOUN (133; 13% instances), SYM (116; 11% instances), ADV (114; 11% instances), VERB (85; 8% instances), PROPN (43; 4% instances), ADJ (41; 4% instances), ADP (39; 4% instances), CCONJ (33; 3% instances), DET (30; 3% instances), PRON (27; 3% instances), SCONJ (24; 2% instances), INTJ (13; 1% instances), NUM (6; 1% instances), X (3; 0% instances)